SDP优化算法：软件定义无线传感器网络中能源消耗最小化资源分配

22 浏览量更新于2024-08-28 收藏 480KB PDF 举报

本文探讨了在软件定义无线传感器网络（SD-WSNs）中通过半定规划（Semidefinite Programming, SDP）进行能源消耗最小化的集中式资源分配算法。研究的核心问题是将节能问题转化为一个优化问题，其中服务质量（Quality of Service, QoS）被量化为信号到干扰加噪声比（Signal-to-Interference-plus-Noise Ratio, SINR）约束。传统的非凸优化问题难以直接求解，因此作者通过将问题放松为半定编程形式，这是一个更为可行的下界估计，提供了有效的求解途径。在理论分析部分，文章引入了两个特殊的案例来深入研究半定规划的下界的有效性。这两个案例可能是针对特定网络结构或应用场景，通过对比实际结果与SDP解，评估了该方法在不同条件下的性能表现。这有助于验证SDP方法在解决实际问题中的实用性。同时，为了进一步探讨资源分配的效率和灵活性，文中还设计了一个分布式方法作为性能基准。相比于集中式方案，分布式方法通常能够减少通信开销和对中心节点的依赖，从而在某些场景下具有优势。通过仿真比较这两种方法，研究者揭示了在SD-WSNs中，何时以及如何选择最适合的资源分配策略以平衡能耗与性能。总结来说，这篇研究论文通过半定规划技术处理了SD-WSNs中的能源优化问题，不仅提出了一个集中式的解决方案，还通过特殊案例分析和分布式方法的引入，提供了全面的理论支持和性能评估。这不仅对能源管理在SD-WSNs中的应用有重要影响，也为未来网络优化策略的设计提供了有价值的参考。

Semideﬁnite Programming based Resource

Allocation for Energy Consumption Minimization

in Software Deﬁned Wireless Sensor Networks

Yueyue Zhang

∗

, Yaping Zhu

∗

, Feng Yan

∗

, Zhengquan Li

∗†

, and Lianfeng Shen

∗

National Mobile Communications Research Laboratory, Southeast University, Nanjing, Jiangsu 210096, P. R. China

†

College of Internet of Things Engineering, Jiangnan University, Wuxi, 214122, China

Email:{douleyue, xyzzyp, feng.yan, lfshen}@seu.edu.cn, lzq722@sina.com

Abstract—In this paper, one centralized resource allocation

algorithm is proposed to minimize energy consumption in soft-

ware deﬁned wireless sensor networks (SD-WSNs). The energy

consumption problem is formulated as an optimization problem,

given quality-of-service (QoS) constraint deﬁned as Signal-to-

Interference-plus-Noise Ratio (SINR). Then, the nonconvex op-

timization problem is relaxed into a semideﬁnite programming

(SDP), which serves as a lower bound. To analyze the tightness of

the lower bound, two special cases are introduced. Besides, one

distributed approach is also developed to provide a performance

benchmark. Furthermore, simulation results are revealed that the

proposed centralized algorithm performances better with respect

to the energy consumption and bandwidth utilization.

Index Terms—resource allocation, software deﬁned wireless

sensor networks (SD-WSNs), semideﬁnite programming (SDP)

I. INTRODUCTION

Software deﬁned networking (SDN) has been a promising

system architecture and control mechanism for future networks

[1]. And the spirit of SDN has been adopted in wireless

domain, for instance, software-deﬁned cognitive radio net-

work [2], software-deﬁned cellular networks [3] and software-

deﬁned VANET [4].

Recent improvement in circuits and wireless communica-

tion technologies makes wireless sensor networks (WSNs)

programmable by integrating signal processing blocks and

other interfaces into a small wireless device. Accordingly,

various researches have been explored to introduce SDN into

WSNs [5], [6]. [5] has proposed software deﬁned WSNs (SD-

WSNs) with Sensor OpenFlow, which synergizes SDN and

WSNs to solve WSN-inherent problems, such as, resource

underutilization, rigid policy changes, and hard management.

In [6], a novel base station architecture realizing SD-WSNs

controller function has been developed, which could help in

tackling several difﬁcult problems in WSNs including energy

saving and network management.

Due to the nature of energy-constrained, sensors in WSNs

are needed to operate without battery replacement or harvested

This work is supported in part by the National Natural Science Foundation

of China (No. 61471164, 61171081, 61571108) and the Research Fund of

National Mobile Communications Research Laboratory, Southeast University

(No. 2016B02)

energy for a long period. In [7], some key design consider-

ations have been addressed for future micro sensor systems,

for example, network protocols, low energy electronics, and

energy harvesting techniques. Besides, optimized modulations

have been studied with the energy constrains when transmit-

ting given number of bits in point-to-point communication [8].

And in [9], minimal energy consumption in WSNs has been

considered in the overall network.

Although some preliminary researches have been done on

SD-WSNs, there seems no resource allocation in SD-WSNs

to reduce energy consumption with the global information

of the whole network. Given QoS constraints, we exploit

resource allocation (power and bandwidth) to minimize energy

consumption in SD-WSNs, which decouples the control logic

from data forwarding. Thus, the information gathered from

sensors will be uploaded to the control plane and could be

processed with a global view of the network.

The energy consumption minimization problem can be

formulate as an optimization problem. However, the non-

convexity makes the search for the global optima hard. Recent-

ly, the semideﬁnite relaxation technique has been studied in

several work [10], [11], which basic idea is to relax the original

non-convex problem into an semideﬁnite programming (SDP).

By taking advantage of the convex optimization technique,

the global minimum of a convex problem can be quickly and

efﬁciently found [10].

Motivated by the above, we propose an SDP-based cen-

tralized algorithm speciﬁcally for the energy consumption

problem in SD-WSNs. Since the original problem is relaxed

into an SDP, it serves as a lower bound , which global optimal

power and bandwidth allocation can be readily obtained.

Besides, the tightness of the lower bound is analyzed by two

proposed special cases. Moreover, an alternative distributed

approach is developed to provide a performance benchmark

of the centralized one.

II. SYSTEM MODEL AND PROBLEM

FORMULATION

A. SD-WSNs Architecture

As depicted in Fig. 1, an SD-WSNs architecture is mainly

divided into two planes, control plane and data plane. In the

control plane, the SDN controllers (SDN-Cs) equipped with

2114

下载后可阅读完整内容，剩余5页未读，立即下载

weixin_38566180

粉丝: 2
资源: 967

SDP优化算法：软件定义无线传感器网络中能源消耗最小化资源分配

Primal-Dual Path-Following Algorithms in Semidefinite Programming

利用半定松弛的高斯混合模型RSS定位WSNs

无线传感器网络定位：LS-SDR与LS-LM算法的性能比较

A filter-sequential semidefinite programming method for nonlinear semidefinite programming

A Robust Adaptive Beamformer Based on Semidefinite Programming with Quadratic Constraints

HANDBOOK OF SEMIDEFINITE PROGRAMMING

Approximation Algorithms and Semidefinite Programming

The application of semidefinite integer programming model for the simulation of complex networks

Handbook of semidefinite programming.pdf

Learning the Kernel Matrix with Semidefinite Programming

最新资源